Automating Knowledge Flows by Extending Conventional Information Retrieval and Workflow Technologies
AdvisorZhao, J. Leon
Committee ChairZhao, J. Leon
MetadataShow full item record
PublisherThe University of Arizona.
RightsCopyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.
AbstractThe efficiency of knowledge flow has been observed to be an important factor in the success of large corporations and communities. In recent years, the concept of knowledge flow has been widely investigated from economics, organizational science and strategic management perspectives. In this dissertation, we study knowledge flows from an Information Technology perspective. The technological challenges to enabling the efficient flow of knowledge can be characterized by two key problems, the passive nature of current knowledge management technologies and the information overload problem.In order to enable efficient flow of knowledge, there is a need for high precision recommender systems and proactive knowledge management technologies that automate knowledge delivery and enable the regulation, control and management of knowledge flows. Although several information retrieval and filtering techniques have been developed over the past decade, delivering the right knowledge to the knowledge workers within the right context remains a difficult problem.In this dissertation, we integrate and build upon the information retrieval and workflow literature to develop and evaluate technologies that address the critical gap in current knowledge management systems. Specifically, we make the following key contributions: (1) we demonstrate a concept-hierarchy-based filtering mechanism and evaluate its efficiency for knowledge distribution. (2) We propose a new architecture that supports the automation of knowledge flow via task-centric document recommendation, and develop a query generation technique for automatically deriving queries from task descriptions and evaluate its efficacy in a domain-specific corpus. (3) We develop an analytical model for predicting the performance of a query and validate the model by analyzing its performance in several domain-specific corpora. (4) We propose a new type of workflow called knowledge workflows to automate the flow of knowledge in an enterprise and present a formal model for representing and executing knowledge workflows.The lack of an enterprise wide knowledge flow infrastructure is one of the major impediments to knowledge sharing across an organization. We believe the technologies proposed in this dissertation will contribute towards a new generation of knowledge management systems that will enable the efficient flow of knowledge and eliminate the technological barriers to knowledge sharing across an organization.
Degree ProgramManagement Information Systems